The three-dimensional structures of a number of human immunodeficiency virus (HIV) protein, glycoprotein or HIV-related molecules are to be determined. These components are likely targets for antiviral agents. It is anticipated that the structures will suggest likely compounds that will viral uncoating, viral attachment or viral penetration. At the same parallel studies are to be conducted on (i) the design of antiviral agents that inhibit uncoating or attachment with respect to viruses whose structures are already known and (ii) structural and mechanistic studies on the neutralization of viruses by antibodies. Not only are these studies quite novel, but they enhance the possibility of designing suitable antiviral compounds and finding neutralizing epitopes once the HIV component molecule's structures are known. A probable antiviral target for HIV is the major capsid protein p24 (program A). It may have a structure similar to many other viral capsids, namely an eight-stranded anti-parallel beta-barrel. It has been shown for rhinoviruses that insertion of a small hydrophobic compound into the interior of the beta-barrel inhibits viral uncoating and that such compounds are, indeed, non-toxic drugs with efficacy. Similar drugs can be designed for p24 both by informed testing prior to the structure determination and by rational design after structure determination. To obtain experience we shall continue to study such drugs bounds to various rhinovirus strains, design drugs for other picornaviruses (of yet unknown structures) (program D1) and attempt to inhibit uncoating for plant and insect viruses with known structure (program D2). The glycoprotein gp120 is the major surface antigen for HIV. This lends itself to numerous structural studies (program B) aimed at investigating neutralization by antibodies and antiviral agents. In particular, compounds might be designed that bind to gp120 and interfere with attachment or fusion. We, therefore, intend to study the structure of both the gp120 (program B) and the CD4 receptor on T lymphocytes (program C). All those compounds which we find can bind to p24 or gp120 in a manner that might cause neutralization will be tested for their efficacy in reducing infectivity of the whole virus in cell culture. Since a knowledge of the structure of gp120 will also reveal the major binding sites for neutralizing antibodies, we also plan to study the mechanism of picornaviral neutralization (program E). This will involve determining the structure of Fab components of neutralizing antibodies which bind to known epitopes on the viral surface.